1 Optimization of Drug Cocktails over Dynamically Changing Landscapes during Cancer Evolution

Closed-loop control of cellular function using a combination of drugs through a stochastic search algorithm was introduced recently on kidney cells with a fixed drug response landscape. However, a variety of cell types including cancer cells, evolve over time and thereby have dynamically changing response landscapes. Driving these cell types to a desired destiny requires optimization of the drug cocktail which minimizes the selection for drug resistance. Here, we perform a feasibility study using a multi-agent stochastic search algorithm to locate optimal drug cocktails for a dynamically changing cellular regime. This study will provide important information on discovering optimal cancer drug cocktails to control tumor cell growth, as well as driving cell fates of other similar cellular systems to desired destinies